The Lujo virus and its deadly trade-off

Viruses are often considered some of the most clever and adaptable entities on the planet. They have evolved countless strategies to outwit the host’s immune defenses, making them challenging opponents in the battle for survival. Among these crafty pathogens is the Lujo virus, a lesser-known member of the Arenaviridae family. Although it hasn’t received as much attention as its cousin, the Lassa virus, Lujo remains a significant concern. This virus has developed a particularly clever way of outsmarting the immune system, which could have substantial implications for future research and the development of antiviral therapeutics.
The Lujo virus was first identified during a small but deadly outbreak in southern Africa in 2008. Although only five people were infected, four of them died due to the lack of effective treatment. Given the lethality of the Lujo virus, the National Institute of Allergy and Infectious Diseases (NIAID) classified this virus as a “Category A Priority Pathogen”. As a zoonotic virus that can easily be transmitted among humans, it poses a significant risk for future pandemics. The possibility that the Lujo virus could be the next deadly virus to cause a pandemic has motivated us to study this dangerous pathogen.
To begin to understand its immune evasion tactics, we decided to take a closer look at the structure of its spike complex, as this complex is responsible for the virus’s ability to infect host cells.

The structure of the Lujo virus spike complex is shown in blue,with the glycan crown around its receptor binding site in orange.Other glycans are colored yellow.

Emerging from the shadows: Unveiling the structure of Lujo virus spike complex
Solving the structure of the Lujo virus spike complex wasn’t straightforward, even though our lab had already succeeded in determining the structure of the spike complex of the genetically related Lassa virus two years ago. Building on the successful purification and data collection strategy we previously developed and following refinement and optimization of the protocol, we finally captured the complete structure of the spike complex of the Lujo virus, including its transmembrane portion, using cryo-electron microscopy (cryo-EM).

Cryo-electron microscopy image of theobtained “top-down” views of the spike,alongside a true top view of the structure.

In our initial cryo-EM experiments, the Lujo virus spike complex consistently adopted a preferred orientation on the grids, resulting in only “top-down” views. This limited distribution of spike orientations made it impossible to get the detailed structural insights we needed. In an effort to obtain ‘side views’ of the spike, we explored multiple approaches, including the change of the purification protocol, the use of other binding proteins, and the deglycosylation of the protein. Yet, the real breakthrough came from an unexpected source: a simple dilution of the sample just before preparing the EM grids. This dilution was originally intended to reduce the glycerol concentration, which helps to stabilize the protein complex, but as a fortunate side effect, it allowed the spike to adopt a broader range of orientations, finally giving us those elusive side views. Once we had success with this method, we set aside all the other experimental strategies and focused on refining the structure we obtained.
Solving the structure of this spike is particularly significant since, unlike other arenaviruses, the Lujo virus does not fall clearly into the classic Old World (OW) or New World (NW) arenavirus classifications. It stands out as a unique pathogen, making the structural resolution of its spike complex a novel contribution to our understanding of arenavirus diversity. Understanding this structure is key to unraveling how the Lujo virus balances its ability to evade the immune system with its need to bind to host cells.
 
The critical balance between evasion and binding
Unlike its relatives in the OW and NW subgroups, which utilize α-dystroglycan and transferrin receptor-1, respectively, as their cellular receptors for gaining cell entry, the highly pathogenic Lujo virus uniquely utilizes neuropilin-2 (NRP2). A few years ago, our lab solved the crystal structure of the receptor binding domain of the Lujo virus in a complex with NRP2. When the structure of the Lassa virus spike complex became available, we compared its receptor binding domain with that of the Lujo virus we had previously crystallized. This comparison suggested that the NRP2 receptor extends outward, potentially allowing space for three monomers to engage with one spike trimer – one on each subunit.
However, determining the structure of the full Lujo virus spike complex revealed an unexpected twist: the relative orientation of the receptor binding domains of the Lujo spike, in the context of the trimer, is different compared to that of the Lassa virus. This change in orientation restricts the trimeric spike to binding with only a single NRP2 monomer, reducing the overall strength of the interaction, as it lacks the avidity that would result from engaging multiple receptors simultaneously within individual spikes.
 

Wild-type and modified Lujo virus spikes (light blue) in complex with the NRP2 receptor (purple).Glycans that interfere with NRP2 receptor binding are shown in magenta.

 
An evolutionary paradox: Modified viruses infect better
One of the most intriguing discoveries we made was that when NRP2 binds to the spike, it is nestled between several glycans that seem to partially interfere with this binding.
Why would a virus hold on to something that prevents it from binding its receptor efficiently?
To investigate further, we engineered a version of the spike that lacked these glycans and assessed its infectivity. Indeed, the spikes without these specific glycans near the receptor binding site improved the ability of the virus to enter cells. This finding suggests that Lujo retains these glycans for a reason. The most likely explanation is that these glycans represent a deliberate adaptation of the virus to improve immune evasion at the expense of compromising the efficiency of its cell entry.
It turns out that the receptor binding domain of the Lujo virus is coated with a glycan crown that acts as a protective shield against the host immune system. This glycan crown helps conceal the receptor binding site from immune recognition by imposing steric constraints. Antibodies that target the receptor binding sites of viruses represent one of the most effective classes of neutralizing antibodies. In the case of the Lujo virus, the immune system will have a hard time mounting an effective response against this site. This glycan shield not only protects the virus from the host immune system but also complicates vaccine development by limiting the accessibility of potential immunogenic sites, highlighting a sophisticated viral strategy to evade host defenses.
 
Take-home message
Binding more tightly to their cellular receptors can make the cell entry process of viruses more efficient, providing them with an evolutionary advantage. For example, once SARS-CoV-2 began spreading in the human population, it quickly acquired mutations that enhanced its affinity for the human ACE2 receptor.
In the case of the Lujo virus, optimizing its affinity for the NRP2 receptor, which is completely conserved among mammals, was clearly not the dominant force that shaped its structure; instead, its spike evolved primarily to evade the immune response. By coating itself with glycans, the virus sacrifices its affinity for NRP2, but in doing so, it becomes incredibly good at staying hidden.
This immune evasion strategy could have been a trait that made the Lujo virus so deadly and dangerous for the people who got infected, as their immune systems couldn’t keep up. The immune responses were probably outpaced, overwhelmed by the virus’s evasion tactics. Elucidating the structure of the Lujo virus spike helps us better understand this virus – and the more we understand it, the better we can fight back.
 
You are very welcome to read the full paper here.
 
Disclaimer: The cover image was generated with the assistance of AI (ChatGPT) and was subsequently polished and colored in Photoshop by the author.